There's No Second Place in Aerospace
Success requires matching equipment capabilities to the demands of the job
By Robert Segal
Vice President and General Manager
Magellan Aerospace USA
Driven by consumers to keep airfares and shipping rates low, aerospace OEMs need to reduce costs from 5 to 10% each year. That directive is passed down through the supply chain. Further, the greatest challenge facing our North American companies is global competition. The largest aircraft customers over the next decade will be China and India, and as a result, many aircraft components that traditionally would be made in Europe, the US, and Canada, are instead being made in these emerging markets. There is also an unusually high demand for titanium, which is impacting us greatly. The material for one of the titanium split-fan cases we produce costs $30,000 per part, compared to $5000 three years ago.
Remaining competitive in aerospace manufacturing, and likely most industries, is a challenge today. One never really wins once and for all time, but skills and results do improve as the bar is raised. That is what continuous improvement in manufacturing is all about—we keep progressing to the next level in process improvement and operational excellence with specific goals as signposts along the way, whether it is a reduction in cycle time, process run time, or manufacturing losses. Fortunately, the technology and lean initiatives we are establishing at Magellan seem to be working. We are still here.
At our Haverhill, MA facility we manufacture and assemble turbine engine support structures, and shafts for commercial and military aircraft. Our 85,000 ft2 (7897 m2) factory can turn, mill, and grind parts as large as 65" (1650-mm) cube with a wide range of machine tools—a mix of about 60% turning and 40% milling.
Structural components support the rotating turbines of the aircraft engine while allowing air to pass through from front to rear. They are generally round, with several struts joining the inner and outer rings to a central bearing housing. Outside support structures may have mounting lugs as attachment points for external engine components, or for mounting the engine to the aircraft.
As in any assembly, each part impacts another. In the event of a turbine component failure, these structures are designed to isolate the problem so that the engine will not affect the body of the aircraft. The structures, like all aircraft components, must withstand a wide range of thermal and g-force conditions. To provide parts that can deal with these conditions, we machine a substantial amount of Inconel 718 and 907 and other nickel alloys for the parts closest to the engine, titanium 6-4 and 6-2-4-2 for its light weight, high strength, and heat-resistant characteristics, and aluminum or magnesium castings for the outer, cooler sections. From a machining viewpoint, one of the challenges we face is achieving high material removal rates while maintaining the form and function of the part.
The bottom line to remaining competitive is the ability to make parts quickly, and that is more than just the speed of the machine tool spindles—it's all of the aspects that need to be in place to bring a product to our customers as quickly as possible, from design through shipping. It's about taking advantage of current technology and applying it correctly.
One of our jobs at this facility is to help our customers design for manufacturability. We don't portray ourselves as designers, but we take advantage of NX (Unigraphics) design tools and automation within software to share our knowledge of fixture design, part families, or part-process families, so that we can plan our operations and setups faster. We also use machine tools that can accommodate the range of materials we use, and the complex part geometries we machine. We have developed the ability to select machine tools in a more standard way, and embark on a comprehensive process when we specify new machinery. We've created a matrix in which we identify our needs and compare those against the relevant machine tool features.
For example, a few years ago we needed to update our machining centers to accommodate larger structural parts that require five-axis machining. We went through an exhaustive survey of every CNC milling machine manufactured. Through our matrix of pluses and minuses, we narrowed the field to three and ultimately selected technology from Mitsui Seiki USA (Franklin Lakes, NJ). Currently we have two of their horizontal five-axis machines with another on order.
These models met all of our criteria, particularly for high-speed contouring with rapid traverse rates of 945 ipm (24 m/min), cutting feeds of 16 m/min, and positioning accuracies of 0.00006" (0.0015 mm) on the larger HU80A-5X model's X, Y, Z envelope and ±2 arcsec on the tilting/rotating A-B-axis trunnion. Of course rigidity and ease of changeover came into play as well, so we can run Inconel, titanium, and magnesium parts on one machine. To be flexible, we need versatile machines.
A key aspect of the machine tool justification process is of course analyzing the cost. One of the great myths in machine tool purchasing is that the price of a piece of equipment is the cost of the equipment, and that is not so. Typically, the price of a machine tool may be as little as 8% of the total life-cycle cost of the machine. Rework, downtime, and maintenance costs increase over time with less-expensive machine tools. Purchasing machines with an effective life cycle of 30,000 operating hours versus machine tools providing 80,000 hours impacts the capital expenditure return significantly.
The characteristics that contribute to a bigger price tag, such as rigidity and precision, FEA enhancements, stiffer spindles and servomotors, and robust CNC technology are among the features we require to machine the new, tougher titaniumnickel alloys, and heavy parts that weigh as much as 1200 kg.
The volumetric accuracy of the Mitsui Seiki machines was one of the aspects that helped give us the positioning accuracies that we need. In high-precision machining of the sort we do, a machine tool's dynamic volumetric accuracy—the location of the machine at any point in the working envelope—should be at least 80% tighter than the part tolerance. If an aircraft gearbox has an overall tolerance of 0.001" (0.03 mm), for example, the machine tool must be capable of true positioning to 0.0002" (0.005 mm) so there is a 0.0008" (0.020-mm) cushion to allow for thermally driven changes and other variances.
Machine geometry is a function of moving a certain amount of mass along the way system. This movement must be straight without pitch, yaw, and roll, because any variation of these three movements is transferred directly to the workpiece. For accurate part production, this concept is very important. Without high volumetric accuracy, our shop would produce unacceptable parts.
Elaborating on this topic, one of our commercial parts is a 43" (1092-mm) diam front frame for an engine used on a regional jet aircraft. It's a magnesium sand casting that we machine complete on the machine. We assemble it here as well. From a material and geometry standpoint, it's not as challenging as, for example, the titanium fan cases we make, but it requires true positioning from 0.001 to 0.002" (0.03*0.05 mm) on a number of different features, which is tight for a part this large. Further, to make the part competitively, we must do it quickly, which is more than just high surface speeds, it requires moving around the part efficiently, fast rapid rates, and quick acc/dec rates. Currently, we're cutting the part dry so there is no oil in the system when we changeover to a titanium part later in the day.
During machining, a probe routine locates the centerline of the part in relation to the HU80A-5X's A and B axes. Next, the inner and outer bolt holes are drilled and tapped using 2/8ths-24 and 5/16-18 straight-flute taps. There are 14 inner holes and 34 outer holes, all 0.375" (9.5 mm) in diam, and requiring a true position of 0.0014" (0.04 mm).
Previously, all three of these operations were done on separate machines. By making the part complete on one machine, we eliminated 19 hr of machine time on each frame part that we produce.
At our company, other initiatives besides advanced design and machining technology are Six Sigma, Kaisen, Kanban, and other lean and quality-oriented approaches. We recently outgrew our facility in Middleton, MA and relocated to a newly refurbished factory in Haverhill. This was an opportunity to completely rethink our plant and office layout. We have maintained a cellular approach on the shop floor, although we rearranged the layout into three part families—Medium Structures, Large Structures, and Shafts. A fourth area is a Support Cell that supports the other cells with fixtures, preset tooling, and gaging. Along with modifying our cells, we restructured our workforce into cell teams, with each cell including a Manufacturing Support Leader who gets product through the cell. This person is supported by a manufacturing engineer and a quality engineer. A Production Control Account Manager schedules material and equipment, and is the liaison with the customer on all the projects that run through a particular cell. To make our transport system safe and more efficient, we installed bridge cranes that cover 75% of the entire shop floor. In the office area, we eliminated the walls to encourage interaction among our technical staff.
How does a part feature want to be cut? In the future, perhaps within the next 15 years, a design feature will be able to carry enough intelligence for its own manufacturing protocols. Much research is being done in the feature-based machining area right now by Mitsui Seiki and other machine tool builders. The machine code that we know today may fade away if this research goes the way it appears to be heading. All machines would have a common language based on features, and the design would carry the information with it.
As manufacturing engineers, we have to understand that technical and economical outputs are coupled. In machining we often talk about the technical output of the system, for example, the microscopic reactions at the cutting zone—what occurs at the tool and part interface. On the other hand, the economic outputs answer the question, "Why am I doing this?" Organizations cannot separate the two. Those of us responsible for machining are faced with newer and more demanding applications and higher performing designs—mostly driven by global economic demands. Thankfully, it's much easier to design complex parts than it was 20 years ago because designers have better tools, such as CAD/CAM systems and analytical tools that permit an increase in feature complexity, pushing the envelope still further.
Although they are more difficult to machine than their predecessors, there are now materials available that permit more demanding applications, such as the new titanium alloys. For years we've been dealing with nickel-based alloys, and today there are even higher-strength nickel alloys created by different processes and manufacturing techniques at the mills. Composite materials are also emerging.
As contract manufacturers in a worldwide market, we have many competitors. And we know that manufacturing technology is transportable. Not only do we have more complex gadgets to make, we must compete with other shops that make the same parts. In this game, innovation and technical skill will enable us to survive.
Magellan Aerospace Corporation
Magellan Aerospace Corp. supplies a wide range of aircraft engine and aerostructure components to aerospace markets, advanced products for military and space markets, and complementary proprietary products for the power, oil, and gas markets. The company employs 3000 people in 17 operating units in Canada, the US, and the UK.
This article was first published in the March 2007 edition of Manufacturing Engineering magazine.
Published Date : 3/1/2007